Pranathi Pashikanti

D

AI/ML Engineer

Datadog, NY, USA

Jul 2025 - Current (1 year 2 months)

• Built Python-based machine learning services for anomaly detection, improving alert prioritization accuracy by 28% across large-scale telemetry datasets • Developed RAG applications using LangChain, OpenAI API, FAISS, and Pinecone, accelerating incident investigation and knowledge retrieval by 35% • Deployed scalable inference APIs using FastAPI, Docker, Kubernetes, and AWS, reducing average response latency by 23% across production services • Automated model deployment pipelines with MLflow and GitHub Actions, decreasing release time by 40% while improving deployment consistency • Monitored production models using Datadog, CloudWatch, and Weights & Biases, reducing unresolved model drift incidents by 26% • Collaborated with backend and p

D

Associate AI/ML Engineer

Datadog, NY, USA

Jan 2025 - Jun 2025 (6 months)

• Engineered NLP classification models using PyTorch and Hugging Face, improving log categorization accuracy by 24% across enterprise monitoring datasets • Built PySpark data pipelines processing telemetry records, reducing feature engineering time by 31% for machine learning model development • Enhanced REST APIs using FastAPI, integrating machine learning services across 12 applications while maintaining 998% service availability • Performed feature engineering using Pandas, NumPy, and SQL, improving anomaly detection precision by 18% across production datasets • Supported CI/CD workflows using GitHub Actions, Docker, and Linux, reducing release validation effort by 27% for AI services • Documented model evaluation results in Jira, improv

D

AI/ML Engineer

DXC Technology, India

Nov 2020 - Dec 2022 (2 years 2 months)

• Developed supervised machine learning models using Python, Scikit-learn, and XGBoost, improving prediction accuracy by 19% across enterprise applications • Built TensorFlow-based classification solutions, reducing manual review effort by 30% through automated prediction workflows • Designed scalable ETL pipelines using Spark, PySpark, and Airflow, increasing enterprise data availability by 34% for analytics teams • Deployed Flask and FastAPI applications on AWS EC2, reliably supporting 120,000+ monthly internal API requests • Managed ML experiments with MLflow, Git, and Docker, reducing model rollback incidents by 22% through controlled versioning • Validated PostgreSQL datasets using SQL, improving reporting accuracy by 22% and reducing

AI/ML Engineer - Intern

Persistent Systems

May 2020 - Oct 2020 (6 months)

• Cleaned and transformed enterprise datasets using Pandas and NumPy, reducing data preparation effort by 25% across multiple projects • Assisted in developing Scikit-learn classification models, improving baseline model accuracy by 11% through feature engineering and hyperparameter tuning • Performed SQL-based data validation on PostgreSQL, identifying 150+ quarterly data quality issues and improving reporting accuracy by 17% • Supported machine learning model testing on Linux environments, reducing deployment defects by 17% through systematic validation • Documented model experiments and sprint updates in Jira, improving development traceability by 25% and reducing onboarding time by 20%

Pranathi Pashikanti

About

Detail

Timeline

Résumé